Awsesome! Looks like end-users have already done the alpha testing and we're now well into the beta testing phase!
I might get by with just adding a fusible link to the test jig and and a better FET! And that diode... and zener on the gate... and... and...
@Robert - Thanks for the pointers. I'll try that along with the recommended zener. In a month or two.
mnem
*tinker-ily*
Keep us posted. I ordered one too
I think that as long as we actually give the FET some protection as Robert sez... we'll be easily able to get more than our money's worth out of it.
The ATORCH listing recommend to
substitute a IRFP260N for full 180W operation, but some of the feedback I see on the unit says it comes with a IRFP260N of dubious origin; either reclaimed or counterfeit.
What I'm seeing so far is that the modders have been subbing single higher voltage and higher continuous current FETs (IXTH130N20T & FDH44N50), adding the base resistor and zener diode to clamp the oscillations from that op-amp, and some suggestions of replacing that TO252 diode with something in a TO247 package. I'm still digging tho.
The listing's 1000W claim is utter fiction; the manufacturer's suggested path to that is to sub a VMO400-02F megaFET module (~$200 list) for the original FET, or to parallel FETs but offers zero recommendations. The latter will still probably be the path I take, however, especially as the IRFP260N is designed with that in mind, at least according to the datasheet.
My desire for this is to have something cheap that can dump 4S regular & HV packs at ~20A discharge to terminal or storage voltage, so ~17volts/300Watts. The specs of the IRFP260N say that is the max dissipation, so I'm guessing at least 2 pass transistors to get there unless I do change to one of the other FETs, then I still need to figure out how to wick the heat away from the FET(s) and that diode fast enough.
Don't think I'm gonna connect any of my big LiPos to it without a 30-ish amp fuse in the test jig, fo sho. Short version: It wasn't obvious at first cuz I'm math-defective; but I think getting more than 10A out of these in my application is going to require some money spent on FETs and maybe a TO-247 Schottky. Nowhere near what a real 300W load would cost, just not the
"Holy fucking asscrackers!" deal this thing looks like on the surface. As is the case with most such cheap, China-direct electronic gadgetry.
I think that if you look at these bare-board units as a somewhat buggy "control module" with serial output to LabView, etc, that you have to build the final pass element of yourself... you'll be going into it with eyes open, and at $28 delivered, that's still a pretty good deal.
mnem